Selenium is an essential trace element with significant health benefits. It has potent cancer prevention activity and roles in the immune system, male reproduction, and many other biological systems and processes. Most of these selenium's effects are mediated by selenocysteine (Sec)-containing proteins. However, information about these proteins is scarce. Sec is known as the 21st amino acid in the genetic code and is co-translationally inserted into proteins in response to UGA codon. Identification and functional analysis of selenoproteins are challenging. Because UGA codon normally serves as a stop signal, selenoprotein genes are misannotated or completely missed in genome annotations. In the previous funding cycle, we developed bioinformatics tools that allow identification of selenoprotein genes by searching for conserved RNA structures, called Sec insertion sequence (SECIS) elements, and identifying selenocysteine/cysteine-containing protein pairs of homologs. By using these tools, we were able to characterize selenoproteomes (entire sets of selenoproteins in an organism) encoded in many completely sequenced genomes, including the human genome. We also provided experimental evidence for the occurrence of newly identified selenoproteins and determined functions of several selenoproteins using computational and experimental approaches. The major focus of the current application is to functionally characterize mammalian selenoproteins. The proposed study will be guided by 3 specific aims: 1) identification of selenoprotein genes in completely sequenced genomes and genome survey projects, demonstration of their occurrence, and analyses of polymorphisms; 2) functional characterization of newly identified mammalian selenoproteins with the focus on their interaction partners; and 3) analysis of selenoprotein functions using model systems. The PI's previous data establish that it is feasible to identify all or almost all selenoprotein genes encoded in completely sequenced genomes and to functionally characterize selenoproteins of particular interest. The results generated should help explain many biological and biomedical effects of dietary selenium, provide insights into evolution of selenocysteine, and identify proteins with redox function. [unreadable] [unreadable] [unreadable]